@inproceedings{zhang-etal-2023-leco,
title = "{LECO}: Improving Early Exiting via Learned Exits and Comparison-based Exiting Mechanism",
author = "Zhang, Jingfan and
Tan, Ming and
Dai, Pengyu and
Zhu, Wei",
editor = "Padmakumar, Vishakh and
Vallejo, Gisela and
Fu, Yao",
booktitle = "Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop)",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.acl-srw.43",
doi = "10.18653/v1/2023.acl-srw.43",
pages = "298--309",
abstract = "Recently, dynamic early exiting has attracted much attention since it can accelerate the inference speed of pre-trained models (PTMs). However, previous work on early exiting has neglected the intermediate exits{'} architectural designs. In this work, we propose a novel framework, Learned Exits and COmparison-based early exiting (LECO) to improve PTMs{'} early exiting performances. First, to fully uncover the potentials of multi-exit BERT, we design a novel search space for intermediate exits and employ the idea of differentiable neural architecture search (DNAS) to design proper exit architectures for different intermediate layers automatically. Second, we propose a simple-yet-effective comparison-based early exiting mechanism (COBEE), which can help PTMs achieve better performance and speedup tradeoffs. Extensive experiments show that our LECO achieves the SOTA performances for multi-exit BERT training and dynamic early exiting.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhang-etal-2023-leco">
<titleInfo>
<title>LECO: Improving Early Exiting via Learned Exits and Comparison-based Exiting Mechanism</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jingfan</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ming</namePart>
<namePart type="family">Tan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pengyu</namePart>
<namePart type="family">Dai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wei</namePart>
<namePart type="family">Zhu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Vishakh</namePart>
<namePart type="family">Padmakumar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gisela</namePart>
<namePart type="family">Vallejo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yao</namePart>
<namePart type="family">Fu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Toronto, Canada</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Recently, dynamic early exiting has attracted much attention since it can accelerate the inference speed of pre-trained models (PTMs). However, previous work on early exiting has neglected the intermediate exits’ architectural designs. In this work, we propose a novel framework, Learned Exits and COmparison-based early exiting (LECO) to improve PTMs’ early exiting performances. First, to fully uncover the potentials of multi-exit BERT, we design a novel search space for intermediate exits and employ the idea of differentiable neural architecture search (DNAS) to design proper exit architectures for different intermediate layers automatically. Second, we propose a simple-yet-effective comparison-based early exiting mechanism (COBEE), which can help PTMs achieve better performance and speedup tradeoffs. Extensive experiments show that our LECO achieves the SOTA performances for multi-exit BERT training and dynamic early exiting.</abstract>
<identifier type="citekey">zhang-etal-2023-leco</identifier>
<identifier type="doi">10.18653/v1/2023.acl-srw.43</identifier>
<location>
<url>https://aclanthology.org/2023.acl-srw.43</url>
</location>
<part>
<date>2023-07</date>
<extent unit="page">
<start>298</start>
<end>309</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T LECO: Improving Early Exiting via Learned Exits and Comparison-based Exiting Mechanism
%A Zhang, Jingfan
%A Tan, Ming
%A Dai, Pengyu
%A Zhu, Wei
%Y Padmakumar, Vishakh
%Y Vallejo, Gisela
%Y Fu, Yao
%S Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop)
%D 2023
%8 July
%I Association for Computational Linguistics
%C Toronto, Canada
%F zhang-etal-2023-leco
%X Recently, dynamic early exiting has attracted much attention since it can accelerate the inference speed of pre-trained models (PTMs). However, previous work on early exiting has neglected the intermediate exits’ architectural designs. In this work, we propose a novel framework, Learned Exits and COmparison-based early exiting (LECO) to improve PTMs’ early exiting performances. First, to fully uncover the potentials of multi-exit BERT, we design a novel search space for intermediate exits and employ the idea of differentiable neural architecture search (DNAS) to design proper exit architectures for different intermediate layers automatically. Second, we propose a simple-yet-effective comparison-based early exiting mechanism (COBEE), which can help PTMs achieve better performance and speedup tradeoffs. Extensive experiments show that our LECO achieves the SOTA performances for multi-exit BERT training and dynamic early exiting.
%R 10.18653/v1/2023.acl-srw.43
%U https://aclanthology.org/2023.acl-srw.43
%U https://doi.org/10.18653/v1/2023.acl-srw.43
%P 298-309
Markdown (Informal)
[LECO: Improving Early Exiting via Learned Exits and Comparison-based Exiting Mechanism](https://aclanthology.org/2023.acl-srw.43) (Zhang et al., ACL 2023)
ACL